1 TB of storage would be nice, but if it means that I have to download 300 GB for a program or a Linux distribution with the same speed of 1 MB/second it would take forever or say a 7 MB web site. We need to see an increase in Internet speeds at affordable prices first before we go overboard with physical storage.
Well, things sadly do tend to move at somewhat the same rate. Linux distributions are bigger than they ever were -- to be fair, they include more than they ever did.
But as for coding, unless RAM gets similarly cheap, you're going to want your programs to be small, because they have to eventually hit RAM. That goes for things like cache size and bus speed, too -- big programs are slow in many ways.
And you seem to be assuming that programmers care more about disk usage than bandwidth usage -- I really doubt that.
But anyway, it would seem to be getting less relevant as fiber gets more common.
Let's assume, for the sake of argument, that a single write takes 100ms.
All you have to do is thread it. You can't do this with normal hard drives, short of RAID. But, like a well-designed flash array, you can pretty much parallize any write you want.
Of course, it means that the filesystem would have to know to do this, but I don't see it really having any serious implications on performance. If it functions as a solid-state device, then I'd assume it could theoretically perform better, actually.
If you think C# is better than C/C++, for arguments sake, there's alwats Java.
I do, first of all -- I get really, really sick of things occasionally segfaulting. It seems to happen more often on amd64. I would understand if it was a low-level library, or something that needs high performance -- something like, say, openssl, but of course, those are already ported and working fine.
Consider that a segmentation fault is basically impossible in C# or Java, or any garbage collected, managed-memory language. Memory leaks are possible, but much easier to avoid. So, every time I see one of those, especially if it's 64-bit related, I silently curse that they weren't written in a higher-level language -- ANY higher-level language, even C#, even Visual Basic.
Don't forget that the differences between C# and Java are tiny.
Let's forget, for the moment, that C# already does have more than Java built in to the language. There is one good thing about the environment -- or bad, depending on how you look at it -- P/Invoke is dirt simple in C#, making bindings to C/C++ libraries much easier. Java makes this significantly harder. Thus, Java apps are much more likely to be completely portable, because no one's going to bother tying it to some platform-specific library unless they have to, and they're far more likely to simply re-implement that library in Java (or download a Java version.)
On the other hand, C# apps (or any.NET app, remember, both.NET and the JVM can support other languages)....NET apps can easily call native libs. This makes it far too easy to implement a Windows-only app in.NET, but it also makes it easy to implement a GNOME-specific app in Mono. With Java, it wouldn't necessarily integrate very well -- wouldn't really feel "native" -- and would be a bitch to integrate where it isn't already.
If Linux needs more specialised languages, there's hundreds of them already.
So why is one more a bad thing?
Pipes aren't a long-term solution, because they only pass serialized data, and they only go one way -- it's hard to set up a more complicated structure, and inefficient for much more than... well, more than what we use it for.
And when it comes to "serious" application development, you almost never see a big desktop Linux app written in anything other than C or C++. Biggest exception I can think of is Eclipse (Java).
Personally, I don't really like Java or.NET right now, but I don't think we've got anything better, for what those are good at, and I certainly don't think Mono is a bad thing.
First, GP had some good points, at least regarding OOXML. They did kind of ruin it with the Mono comment, though.
Second, you mention Java, PHP, and Perl -- how much have you used Perl? What about Python or Ruby?
I can say that while Visual Studio is nice, I've been sticking to eclipse, and in both, there are things I really miss about my old vim environment. I have to say I was most productive, oddly enough, on a Mac, running Terminal, ssh'd in to a Linux machine, using vim and ruby.
Interoperability is not a bad thing. However, using them instead of what we believe to be better tools, locally, is a bad thing. Unless he's planning to develop Gnome components in Visual Studio on Windows, you don't really get the.NET development environment. Being tied to a commercial MS product is definitely a bad thing.
If Samba didn't exist, it would be a bit more hassle, but not much -- we have WebDAV, FTP, NFS (I'm fairly sure there's a Windows client somewhere), SCP/SFTP, rsync, and so on.
There was not read/write support for NTFS for most of the time I've been using Linux. The write support came recently, and was still quite painful to do until very recently (this year or so). For awhile, the read support wasn't necessarily dependable either, so we would either install Windows on FAT, or have a separate FAT partition -- or use Samba servers. And Linux still makes a great server -- arguably a better webserver -- without any of these things.
Lastly, I don't really care if Linux is a more competitive OS. I only want it to be a capable OS. I realize it has to be a bit of both, but for the most part, I can actually live without Windows, if I have to. Thus, it's Windows (Vista, especially) which needs to prove itself "competitive", at least to me.
I used to have a machine very similar in use to this thing, though it cost quite a bit more. It had a 15 gig hard drive (actual, spinning, iPod-like drive) and a 1 ghz processor.
First, that's probably not enough to play movies. Maybe it was the video card, too, but while it was fine for fansubs (anime), it sucked for actual DVDs.
But second, I'd recommend taking stuff on the hard drive, anyway. Load some 5 gigs worth of pre-encoded stuff (and let a desktop do the encoding), or get some more memory and rip the whole DVD.
The last thing you want to spend battery life on -- any battery -- is spinning a huge plastic disc.
jffs2 runs on a raw flash device. It won't work on a normal block device, or hard disk, etc. There's an emulation layer that could make it work, but there'd be no point.
Just guessing here, but I think you'd want jffs2 if the flash supports being read as flash by Linux, and some other filesystem (ext3 or something else, but that's another flamewar) if it appears as a disk.
Assuming this is cheap, you just buy another one to back it up on. (And if it's not cheap, you probably can't afford the first one, anyway.)
The bigger problem is a social engineering one. Someone is going to forget to backup, and someone is going to get their data stolen. But you can't solve these with technology any more than you can cause them with technology.
I do have to admit that it would be silly of them to include SSH/SFTP by default, but this is what every other OS is able to do, at least for X apps -- ssh with X forwarding.
Not that there are that many GUI tools to make you want it -- and most of the GUI admin tools you'd care about are either a web interface or already provide their own client/server model, thus making it possible to admin them via the same native interface on your own Linux desktop.
Dogma implies that people of faith are following something merely because it is pushed by a church and hammered into their skulls, not that people are capable of independent thought and coming to their own conclusions.
Well, much of the scientific community has a dogma suggesting that anyone capable of independent thought who actually takes the time to use it will come to the same conclusion on at least a few irrefutable truths.
In other words, creationism is false, and if you believe it, you're either dogmatic or deluded.
I would tend to agree with this in most cases, but what really needs to happen here (and I may not be capable of it) is to accept that someone is not necessarily stupid or insane, or even wrong, simply because they disagree with you.
But I have to admit -- I have yet to see an argument for creationism that isn't dogmatic. The best most people can come up with is something along the lines of "Because the Bible says so", or "Because I want to", or "Because you can't prove it didn't happen (thus implying faith is the default position)"...
Still, if you have anything better, I'm willing to listen. If all you can do is call us childish -- call us names -- well, then, familiarity breeds contempt.
Not everywhere, but in enough places that I'd think people would notice.
For example: The Linux kernel. I'm running 2.6.22. The 2 is most likely incrementing normally, since there was a 1.0.0 release, that was considered "stable", or as much as it can be. The 22 also increments normally, I think -- though I may be wrong about that.
But I did upgrade directly from 2.4 to 2.6. This is because Linux 2.5 was a development branch. Highly unstable, but it went on for quite awhile, with the most essential parts backported to 2.4. When it was stable enough, 2.6 was released, starting with 2.6.0_rc1 (I think) -- but 2.4 is still maintained, maybe even 2.2 (or did they finally drop that?)
An under-used server is a waste, I'll agree, provided you have more than one of them. That's because one server at 50% CPU is likely consuming less power than two servers at 20% CPU each. Even the CPU is probably more efficient, but there's the rest of the box to consider.
But, desktops? Hell no. Aside from reliability issues -- figuring out which ones are on and available, and where to route a request -- there's security -- just who's computer gets to process your credit card information? What if it's not malicious -- how do you detect bad RAM/disk/network/CPU? (Yes, CPU -- think Pentium bug.)
SETI manages to deal with all of that by running many jobs on more than one machine, so it can verify that either they are all lying the same way, or they aren't cheating -- that, and it's a closed source program.
But I don't run SETI, or Folding@Home, or Distributed.net, and I wouldn't run this.
The reason is simple: Try anything that uses a lot of CPU on a laptop. My first laptop got a little warm, and worried me -- but it was passively cooled, and I could just give it a little air. My Powerbook would get hot, but not particularly loud. My newest laptop is a Toshiba, and when it idles, it still occasionally cranks the fan up to full every few minutes to blast some very hot air out the side. When I crank the CPU(s), it could turn on most or all of the time.
All of them would use significantly more battery life when I was using the CPU. The first one was the worst, it would die in 2-3 hours instead of 9-12 (!).
Anyway, if that's the power difference on a laptop, imagine on a desktop -- not to mention the noise. No thanks.
That's actually what I loved about the first one -- a Sharp MM10 -- it would use almost no power, and weighed almost nothing, yet I could use it as a dumb terminal, connect to my server, get some real work done.
So... no. Not unless you are willing to make my desktop quiet enough, and pay me for the power usage, and not complain when I steal all your private data -- but if you're willing to pay, why not buy it from Amazon's S3, or this Cisco thing, or something similar?
But it's more like a collection of virtual machines, and a collection of servers. You buy the virtual machines, they figure out where to put em -- but nowhere near as flexible as a real cluster would be.
Yeah, I think the key thing here is that Greenpeace has an end goal of getting attention. Once they get attention, then their goal is to say their message.
Then they have an end goal of saying their message. Or maybe of saying their message to a large enough, attentive group of people.
I'm not even sure that's their end goal, that's just what I can derive from your claim. If I were forming such a group, I'd have an end goal of actually making the world change, of protecting the environment and helping it to flourish... Not merely of "raising awareness" or some other bullshit that has a declaration of defeat right there in the goal.
After all, if I could solve pollution, say, why would I want to raise awareness about it? And if my goal is only to raise awareness, doesn't that mean that if a few hundred million people tell me to fuck off and die in a fire, I've attained my goal?
But maybe that's the fundamental flaw here -- maybe Greenpeace has really forgotten what their goal is, and has settled merely for raising awareness. Or maybe they've added "by any means necessary" -- or both.
However, they have to get attention, and so they do stuff like this
I know that conventional wisdom is "no news is bad news". But I don't think that applies here -- stories like this might be better than no one knowing what the word "environmentalist" means, but in the long run, far worse than taking some extra time and money and doing it right.
There is no standard way of monitoring RAID/Fans/Hardware failures etc.
Google for lm-sensors. I'm fairly sure that's at least a standard API, even if the backends aren't standard.
There's also SNMP and Nagios, which can be used to remotely monitor a system. I'm fairly sure you can tie these in to lm-sensors.
Each vendor has their own tools which makes having multi-vendor environments a pain, If we compare against windows with mom every vendor has a plugin which will allow you to monitor and manage the systems from a central point.
HAHAHAHAHAHAHAHA....
Oh, man... wait... you were serious?
Just look at a typical Windows laptop. I have a tool for my touchpad, which is relatively standard only because it seems like Synaptic makes all laptop touchpads. I have a tool for my thumbprint scanner, a tool to update the Toshiba drivers, and while I've opted to use the normal XP wireless support (I think), there's that plus an Intel wireless utility... Even soundcards frequently come with their own system tray thing.
Now, maybe it's different in the enterprise, but it seems to me that hardware vendors are always creating their own little utilities before there's a standard, and are slow to adopt the standard once it exists, for any OS except OS X, and maybe Solaris (before Solaris/x86 and OpenSolaris).
I'll give you the fibre, mostly because I don't know anything about that, but then, consider that Google has an infrastructure basically built out of commodity desktop hardware. So, if you have to, you can always just throw known hardware configurations at it -- which is really what you should do with any OS.
First, this has been tried, and while the GPL deliberately and specifically allows this, some armchair lawyers turned armchair crackers kept breaking in and taking the code back.
Second, even if it were done, all that means is the community needs one person to buy the product, open the CD/DVD (or encrypted archive, get into this century already), and upload it to a public FTP (or HTTP/WebDAV, get into this century already). Remember, it comes with the GPL, free or not, which means a license to redistribute.
Yes, it could get expensive, but the only point at which this becomes entirely closed is if it is GPLv2 (not v3) and used for a service, like a website. There are likely a few similar loopholes such that a company can sell proprietary, GPL code, and be fairly confident that it won't reach the Internet.
Then again, I would hope that the fairest, most likely way to get paid to develop GPL code is by the feature, not by the license. That is, someone pays you to take some existing GPL'd software and add the functionality they need. In general, corporations are competing on things completely unrelated to the software they use, so they could care less if their competitors get to use the features they pay for -- so much the better, if said competitors also improve said software.
Sure, you could keep those improvements to yourself, but the cost of keeping your own private set of patches up to date with the latest development is prohibitive, when compared to simply letting the community (including you and your competitors) maintain it as a single tree.
So basically facebook changed there maximum users from a huge number to an even bigger number.
The difference being that the current (huge) number was not quite sufficient to register all human beings on the planet, so we have to wonder why they did this. 32-bit integers are kind of the default, so most people wouldn't worry about it. So why are they doing this?
If you look, this article is filed under "It's funny. Laugh." And it is, really. Either Facebook is doing this for no good reason, or someone actually has some justification for going to the time and expense to change their database in this way. And so... Are they really planning on registering more human beings that exist?
I realize it's not funny to you now, as you've had to have the joke explained to you...
Are we going to post a news story everytime google adds to their storage system?
If they suddenly went from 2 gigs of email to 5 exabytes, then yes.
Also, keep in mind that Slashdot did cover when Gmail was first released with that 2 gigs, which seemed impossibly huge, and was at least one or two orders of magnitude larger than their closest competitor.
Other consoles could have the Wiimote, but they still wouldn't have Nintendo's attitude towards games and would probably not market them in the same way that Nintendo have. It's not just about the hardware or the software, but about the way that they've been put together in a cohesive package and
Hmm... and apparently you can't finish that thought.
Let me finish it for you: Anyone can do a "cohesive package", on any platform, open or not. Steam, on the PC, is a nice, one-stop, "just works" platform for gaming, like Xbox Live -- but you can do PC gaming without Steam, and you can do other things with your PC.
What I don't like is that consoles are multiple, closed platforms, each controlled by a monopoly. Nintendo makes good games, sure, but because they control the Wii, they get to decide whether, for instance, it supports high-def. They can make more specialized hardware inside the box, too, to make it cheaper, but if I'm willing to buy $1000 worth of computer on which to play games, I still won't be able to use it to make Wii games look any better.
Right now, for a developer to support every platform out there is somewhat difficult. But for a gamer to be able to play any game out there is downright impossible. You justify it by implying that each console comes with an Experience (TM), and that some people like some experiences, and other people like others. Well, I like shooters, strategy, platform games, Zelda and the Wiimote, and fl0w. Why should I have to buy a $1k computer, a $200 console, a $400 console, and a $600 console, just to be able to play any game I want?
Why can't I just buy a $1k console, and have that be the end of it for another five years?
I said it was the platform of choice for the hardcore Shooter, Stratergy and MMO markets.
You do know it's spelled "strategy", right?
And no, it's not. Strategy and MMO, maybe. Shooter, well, maybe it's just me, but almost everyone I know who has played PC games has also played Halo, and I can't say the reverse is true.
However, to renforce my earlier point the hardcore shooter market will want private clan servers and modifications. This is somthing that you can't get with Live.
Mods, no. Private servers, yes, easily. Have you even tried?
And why do "hardcore" n00bs with l33t mods get consideration, but not professional gamers?
the topic isnt new, as much as you want to nay say it, the possibility is there.
I know. There are even a few out there. Gentoo put out a LiveCD which had America's Army, for instance.
And that no longer works -- the drivers don't support newer video cards. Which is fine, though annoying, for America's Army -- you can just download another one. But it sucks for a game you paid for, and it still sucks for America's Army -- why download another 700 meg livecd (or 4.3 gigabyte DVD) to patch some 5-10 megs worth of drivers?
And you're right, it's not new. There were a couple of old DOS games which shipped on a bootable CD, maybe even a bootable floppy. All of which are painful to play now, because of the same reason -- no hardware support. Plus, a modern computer could play probably ten or twenty copies of those games, simultaneously, if they didn't depend on booting this specialized OS.
It wouldn't even work for new games. The day Doom3 came out, both nVidia and ATI had new drivers, with improvements for OpenGL.
I am not saying that you can't make a game that boots from a DVD. I'm saying you'd have to be retarded to try it, when it's so much less work simply to port it to the OS. If it's really such a big deal, write your game in Java -- then you won't even have to recompile for other CPU architectures, let alone OSes.
all work in various forms, yes some have had problems, but so have most o/s's so that is a pretty dubious argument.
Most OSes can at least patch themselves.
one could chop out a lot of cruft and specialize the games os to run full out for gaming, not to mention reduce the memory footprint
Which has nothing to do with livecds.
Also, memory is hardly the bottleneck. CPU, video card usage, network performance, etc, these things matter, but they are also not something that the OS (besides Vista) will interfere with.
the developing company would have complete control over the games os...this is a big advantage over having to integrate with an os you dont have control over.
And a big disadvantage as soon as you want -- oh -- a savegame. Then you have to integrate with a filesystem you don't have control over, without the benefit of the OS drivers for it.
Let's not forget the cost of rebooting. Ever notice how long it takes to load a livecd? Now consider that you have to stop everything else the computer is capable of...
I would very much like to leave my torrents running while I play games, thanks.
You will never reach a wide audience with a "1 size fits all" console, because 1 size never fits all.
Tell that to the people using the PC, for anything other than gaming. The biggest problem now is software -- one OS doesn't fit all -- but one hardware platform certainly does.
And that "one OS doesn't fit all" is largely a result of existing OSes not cooperating on a standard, the way the hardware manufacturers do.
However, most of your reasons are pure bullshit.
Wii has the Wiimote -- why shouldn't that be able to plug into any console? Or into a PC?
PC is for the hardcore -- bullshit. I'd say people who make six figures winning Halo tourneys are pretty hardcore.
Xbox is doing what it was made to do -- which is what, exactly?
And why should the PS3 be different than the PC, exactly? The only real advantage here is price and exclusivity, which are, again, bullshit reasons.
There is no technical reason why this can't happen. All the reasons are bullshit financial/business reasons.
Slashdot Mirror
Well, things sadly do tend to move at somewhat the same rate. Linux distributions are bigger than they ever were -- to be fair, they include more than they ever did.
But as for coding, unless RAM gets similarly cheap, you're going to want your programs to be small, because they have to eventually hit RAM. That goes for things like cache size and bus speed, too -- big programs are slow in many ways.
And you seem to be assuming that programmers care more about disk usage than bandwidth usage -- I really doubt that.
But anyway, it would seem to be getting less relevant as fiber gets more common.
Let's assume, for the sake of argument, that a single write takes 100ms.
All you have to do is thread it. You can't do this with normal hard drives, short of RAID. But, like a well-designed flash array, you can pretty much parallize any write you want.
Of course, it means that the filesystem would have to know to do this, but I don't see it really having any serious implications on performance. If it functions as a solid-state device, then I'd assume it could theoretically perform better, actually.
Or Enlightenment? Or Fluxbox? Or WindowMaker?
Really?
I do, first of all -- I get really, really sick of things occasionally segfaulting. It seems to happen more often on amd64. I would understand if it was a low-level library, or something that needs high performance -- something like, say, openssl, but of course, those are already ported and working fine.
Consider that a segmentation fault is basically impossible in C# or Java, or any garbage collected, managed-memory language. Memory leaks are possible, but much easier to avoid. So, every time I see one of those, especially if it's 64-bit related, I silently curse that they weren't written in a higher-level language -- ANY higher-level language, even C#, even Visual Basic.
Let's forget, for the moment, that C# already does have more than Java built in to the language. There is one good thing about the environment -- or bad, depending on how you look at it -- P/Invoke is dirt simple in C#, making bindings to C/C++ libraries much easier. Java makes this significantly harder. Thus, Java apps are much more likely to be completely portable, because no one's going to bother tying it to some platform-specific library unless they have to, and they're far more likely to simply re-implement that library in Java (or download a Java version.)
On the other hand, C# apps (or any .NET app, remember, both .NET and the JVM can support other languages)... .NET apps can easily call native libs. This makes it far too easy to implement a Windows-only app in .NET, but it also makes it easy to implement a GNOME-specific app in Mono. With Java, it wouldn't necessarily integrate very well -- wouldn't really feel "native" -- and would be a bitch to integrate where it isn't already.
So why is one more a bad thing?
Pipes aren't a long-term solution, because they only pass serialized data, and they only go one way -- it's hard to set up a more complicated structure, and inefficient for much more than... well, more than what we use it for.
And when it comes to "serious" application development, you almost never see a big desktop Linux app written in anything other than C or C++. Biggest exception I can think of is Eclipse (Java).
Personally, I don't really like Java or .NET right now, but I don't think we've got anything better, for what those are good at, and I certainly don't think Mono is a bad thing.
First, GP had some good points, at least regarding OOXML. They did kind of ruin it with the Mono comment, though.
.NET development environment. Being tied to a commercial MS product is definitely a bad thing.
Second, you mention Java, PHP, and Perl -- how much have you used Perl? What about Python or Ruby?
I can say that while Visual Studio is nice, I've been sticking to eclipse, and in both, there are things I really miss about my old vim environment. I have to say I was most productive, oddly enough, on a Mac, running Terminal, ssh'd in to a Linux machine, using vim and ruby.
Interoperability is not a bad thing. However, using them instead of what we believe to be better tools, locally, is a bad thing. Unless he's planning to develop Gnome components in Visual Studio on Windows, you don't really get the
If Samba didn't exist, it would be a bit more hassle, but not much -- we have WebDAV, FTP, NFS (I'm fairly sure there's a Windows client somewhere), SCP/SFTP, rsync, and so on.
There was not read/write support for NTFS for most of the time I've been using Linux. The write support came recently, and was still quite painful to do until very recently (this year or so). For awhile, the read support wasn't necessarily dependable either, so we would either install Windows on FAT, or have a separate FAT partition -- or use Samba servers. And Linux still makes a great server -- arguably a better webserver -- without any of these things.
Lastly, I don't really care if Linux is a more competitive OS. I only want it to be a capable OS. I realize it has to be a bit of both, but for the most part, I can actually live without Windows, if I have to. Thus, it's Windows (Vista, especially) which needs to prove itself "competitive", at least to me.
I used to have a machine very similar in use to this thing, though it cost quite a bit more. It had a 15 gig hard drive (actual, spinning, iPod-like drive) and a 1 ghz processor.
First, that's probably not enough to play movies. Maybe it was the video card, too, but while it was fine for fansubs (anime), it sucked for actual DVDs.
But second, I'd recommend taking stuff on the hard drive, anyway. Load some 5 gigs worth of pre-encoded stuff (and let a desktop do the encoding), or get some more memory and rip the whole DVD.
The last thing you want to spend battery life on -- any battery -- is spinning a huge plastic disc.
jffs2 runs on a raw flash device. It won't work on a normal block device, or hard disk, etc. There's an emulation layer that could make it work, but there'd be no point.
Just guessing here, but I think you'd want jffs2 if the flash supports being read as flash by Linux, and some other filesystem (ext3 or something else, but that's another flamewar) if it appears as a disk.
We already have all kinds of good encryption.
Assuming this is cheap, you just buy another one to back it up on. (And if it's not cheap, you probably can't afford the first one, anyway.)
The bigger problem is a social engineering one. Someone is going to forget to backup, and someone is going to get their data stolen. But you can't solve these with technology any more than you can cause them with technology.
I do have to admit that it would be silly of them to include SSH/SFTP by default, but this is what every other OS is able to do, at least for X apps -- ssh with X forwarding.
Not that there are that many GUI tools to make you want it -- and most of the GUI admin tools you'd care about are either a web interface or already provide their own client/server model, thus making it possible to admin them via the same native interface on your own Linux desktop.
Well, much of the scientific community has a dogma suggesting that anyone capable of independent thought who actually takes the time to use it will come to the same conclusion on at least a few irrefutable truths.
In other words, creationism is false, and if you believe it, you're either dogmatic or deluded.
I would tend to agree with this in most cases, but what really needs to happen here (and I may not be capable of it) is to accept that someone is not necessarily stupid or insane, or even wrong, simply because they disagree with you.
But I have to admit -- I have yet to see an argument for creationism that isn't dogmatic. The best most people can come up with is something along the lines of "Because the Bible says so", or "Because I want to", or "Because you can't prove it didn't happen (thus implying faith is the default position)"...
Still, if you have anything better, I'm willing to listen. If all you can do is call us childish -- call us names -- well, then, familiarity breeds contempt.
I'd like to file a bug report.
Not everywhere, but in enough places that I'd think people would notice.
For example: The Linux kernel. I'm running 2.6.22. The 2 is most likely incrementing normally, since there was a 1.0.0 release, that was considered "stable", or as much as it can be. The 22 also increments normally, I think -- though I may be wrong about that.
But I did upgrade directly from 2.4 to 2.6. This is because Linux 2.5 was a development branch. Highly unstable, but it went on for quite awhile, with the most essential parts backported to 2.4. When it was stable enough, 2.6 was released, starting with 2.6.0_rc1 (I think) -- but 2.4 is still maintained, maybe even 2.2 (or did they finally drop that?)
An under-used server is a waste, I'll agree, provided you have more than one of them. That's because one server at 50% CPU is likely consuming less power than two servers at 20% CPU each. Even the CPU is probably more efficient, but there's the rest of the box to consider.
But, desktops? Hell no. Aside from reliability issues -- figuring out which ones are on and available, and where to route a request -- there's security -- just who's computer gets to process your credit card information? What if it's not malicious -- how do you detect bad RAM/disk/network/CPU? (Yes, CPU -- think Pentium bug.)
SETI manages to deal with all of that by running many jobs on more than one machine, so it can verify that either they are all lying the same way, or they aren't cheating -- that, and it's a closed source program.
But I don't run SETI, or Folding@Home, or Distributed.net, and I wouldn't run this.
The reason is simple: Try anything that uses a lot of CPU on a laptop. My first laptop got a little warm, and worried me -- but it was passively cooled, and I could just give it a little air. My Powerbook would get hot, but not particularly loud. My newest laptop is a Toshiba, and when it idles, it still occasionally cranks the fan up to full every few minutes to blast some very hot air out the side. When I crank the CPU(s), it could turn on most or all of the time.
All of them would use significantly more battery life when I was using the CPU. The first one was the worst, it would die in 2-3 hours instead of 9-12 (!).
Anyway, if that's the power difference on a laptop, imagine on a desktop -- not to mention the noise. No thanks.
That's actually what I loved about the first one -- a Sharp MM10 -- it would use almost no power, and weighed almost nothing, yet I could use it as a dumb terminal, connect to my server, get some real work done.
So... no. Not unless you are willing to make my desktop quiet enough, and pay me for the power usage, and not complain when I steal all your private data -- but if you're willing to pay, why not buy it from Amazon's S3, or this Cisco thing, or something similar?
They have a cloud like this, running on Xen.
But it's more like a collection of virtual machines, and a collection of servers. You buy the virtual machines, they figure out where to put em -- but nowhere near as flexible as a real cluster would be.
What the fuck is that, seriously?
Are we to stream video from a server somewhere that has a beefy video card?
Because if so, that's the dumbest idea I've heard in awhile, though it may make cheating harder.
If not, I don't see what you mean by a "virtual client" or what it has to do with this concept.
Then they have an end goal of saying their message. Or maybe of saying their message to a large enough, attentive group of people.
I'm not even sure that's their end goal, that's just what I can derive from your claim. If I were forming such a group, I'd have an end goal of actually making the world change, of protecting the environment and helping it to flourish... Not merely of "raising awareness" or some other bullshit that has a declaration of defeat right there in the goal.
After all, if I could solve pollution, say, why would I want to raise awareness about it? And if my goal is only to raise awareness, doesn't that mean that if a few hundred million people tell me to fuck off and die in a fire, I've attained my goal?
But maybe that's the fundamental flaw here -- maybe Greenpeace has really forgotten what their goal is, and has settled merely for raising awareness. Or maybe they've added "by any means necessary" -- or both.
I know that conventional wisdom is "no news is bad news". But I don't think that applies here -- stories like this might be better than no one knowing what the word "environmentalist" means, but in the long run, far worse than taking some extra time and money and doing it right.
Google for lm-sensors. I'm fairly sure that's at least a standard API, even if the backends aren't standard.
There's also SNMP and Nagios, which can be used to remotely monitor a system. I'm fairly sure you can tie these in to lm-sensors.
HAHAHAHAHAHAHAHA....
Oh, man... wait... you were serious?
Just look at a typical Windows laptop. I have a tool for my touchpad, which is relatively standard only because it seems like Synaptic makes all laptop touchpads. I have a tool for my thumbprint scanner, a tool to update the Toshiba drivers, and while I've opted to use the normal XP wireless support (I think), there's that plus an Intel wireless utility... Even soundcards frequently come with their own system tray thing.
Now, maybe it's different in the enterprise, but it seems to me that hardware vendors are always creating their own little utilities before there's a standard, and are slow to adopt the standard once it exists, for any OS except OS X, and maybe Solaris (before Solaris/x86 and OpenSolaris).
I'll give you the fibre, mostly because I don't know anything about that, but then, consider that Google has an infrastructure basically built out of commodity desktop hardware. So, if you have to, you can always just throw known hardware configurations at it -- which is really what you should do with any OS.
First, this has been tried, and while the GPL deliberately and specifically allows this, some armchair lawyers turned armchair crackers kept breaking in and taking the code back.
Second, even if it were done, all that means is the community needs one person to buy the product, open the CD/DVD (or encrypted archive, get into this century already), and upload it to a public FTP (or HTTP/WebDAV, get into this century already). Remember, it comes with the GPL, free or not, which means a license to redistribute.
Yes, it could get expensive, but the only point at which this becomes entirely closed is if it is GPLv2 (not v3) and used for a service, like a website. There are likely a few similar loopholes such that a company can sell proprietary, GPL code, and be fairly confident that it won't reach the Internet.
Then again, I would hope that the fairest, most likely way to get paid to develop GPL code is by the feature, not by the license. That is, someone pays you to take some existing GPL'd software and add the functionality they need. In general, corporations are competing on things completely unrelated to the software they use, so they could care less if their competitors get to use the features they pay for -- so much the better, if said competitors also improve said software.
Sure, you could keep those improvements to yourself, but the cost of keeping your own private set of patches up to date with the latest development is prohibitive, when compared to simply letting the community (including you and your competitors) maintain it as a single tree.
For an xbox, maybe, but there's still a significant disadvantage.
The difference being that the current (huge) number was not quite sufficient to register all human beings on the planet, so we have to wonder why they did this. 32-bit integers are kind of the default, so most people wouldn't worry about it. So why are they doing this?
If you look, this article is filed under "It's funny. Laugh." And it is, really. Either Facebook is doing this for no good reason, or someone actually has some justification for going to the time and expense to change their database in this way. And so... Are they really planning on registering more human beings that exist?
I realize it's not funny to you now, as you've had to have the joke explained to you...
If they suddenly went from 2 gigs of email to 5 exabytes, then yes.
Also, keep in mind that Slashdot did cover when Gmail was first released with that 2 gigs, which seemed impossibly huge, and was at least one or two orders of magnitude larger than their closest competitor.
Would've been funnier if it had anything at all to do with 64-bit.
Hmm... and apparently you can't finish that thought.
Let me finish it for you: Anyone can do a "cohesive package", on any platform, open or not. Steam, on the PC, is a nice, one-stop, "just works" platform for gaming, like Xbox Live -- but you can do PC gaming without Steam, and you can do other things with your PC.
What I don't like is that consoles are multiple, closed platforms, each controlled by a monopoly. Nintendo makes good games, sure, but because they control the Wii, they get to decide whether, for instance, it supports high-def. They can make more specialized hardware inside the box, too, to make it cheaper, but if I'm willing to buy $1000 worth of computer on which to play games, I still won't be able to use it to make Wii games look any better.
Right now, for a developer to support every platform out there is somewhat difficult. But for a gamer to be able to play any game out there is downright impossible. You justify it by implying that each console comes with an Experience (TM), and that some people like some experiences, and other people like others. Well, I like shooters, strategy, platform games, Zelda and the Wiimote, and fl0w. Why should I have to buy a $1k computer, a $200 console, a $400 console, and a $600 console, just to be able to play any game I want?
Why can't I just buy a $1k console, and have that be the end of it for another five years?
You do know it's spelled "strategy", right?
And no, it's not. Strategy and MMO, maybe. Shooter, well, maybe it's just me, but almost everyone I know who has played PC games has also played Halo, and I can't say the reverse is true.
Mods, no. Private servers, yes, easily. Have you even tried?
And why do "hardcore" n00bs with l33t mods get consideration, but not professional gamers?
Dance Dance Revolution.
One game is all you need, if it's good enough.
Tell that to the people using the PC, for anything other than gaming. The biggest problem now is software -- one OS doesn't fit all -- but one hardware platform certainly does.
And that "one OS doesn't fit all" is largely a result of existing OSes not cooperating on a standard, the way the hardware manufacturers do.
However, most of your reasons are pure bullshit.
Wii has the Wiimote -- why shouldn't that be able to plug into any console? Or into a PC?
PC is for the hardcore -- bullshit. I'd say people who make six figures winning Halo tourneys are pretty hardcore.
Xbox is doing what it was made to do -- which is what, exactly?
And why should the PS3 be different than the PC, exactly? The only real advantage here is price and exclusivity, which are, again, bullshit reasons.
There is no technical reason why this can't happen. All the reasons are bullshit financial/business reasons.